Shulchan Aruch: Chapter 32 - Laws Relating to Writing of Tefillin

SECTION 32 Laws Relating to the Writing of Tefillin. (1–80)

1 The1 mitzvah of tefillin [requires] that one write out the four passages in which this mitzvah is mentioned, i.e., Kadesh li chol bechor until lemoadah miyamim yamimah,2 Vehayah ki yeviacha until ki bechozek yad hotzianu HaShem miMitzrayim,3 Shema until uvishe’arecha,4 and Vehayah im shamoa until al haaretz.5

These passages must be written in the order in which they appear in the Torah;6 i.e., first the passage that begins Kadesh, then Vehayah ki yeviacha, then Shema, and then Vehayah im shamoa. If one changed the order, (tefillin which use these passages) are invalid.7 (The passages themselves, however, may be used, as follows: If a passage which appears earlier in the Torah was written after other passages, one may write the later passages afresh and combine them into a set.8 Similarly, the passages which appear later [in the Torah] but which were written first can be combined with an earlier passage taken from other tefillin whose later passages had become invalid.9 If [these passages that were written out of order] cannot be combined [with others], they must be set aside.10)

Some authorities maintain that it is also fitting [to write] the passages of the arm-tefillin before those of the head-tefillin, because they are mentioned first in the verse.11 Nevertheless, if one does not [write these passages] first, [the tefillin] are not invalidated. (Others maintain that, as an initial preference, it is proper [to write] the passages of the head-tefillin before those of the arm-tefillin, because the holiness of the head-tefillin is greater than that of the arm-tefillin.12 This is proper according to the teachings of the Kabbalah.)13

2 According to the teachings of the Kabbalah, the dalet of the word echad should be written four times larger than a small dalet.14

3 The initial and preferred option is that one write the four passages of the arm-tefillin on one sheet of parchment.15 This is indicated by the phrase,16 “And it shall be for you as a sign,” [whose singular form] implies [that there should be] one sign, i.e., one compartment, visible as a sign from the outside.17 And just as [the tefillin] are a single sign on the outside, so too, should there be a single sign on the inside; i.e., all the passages should [be written] as a single epistle on one sheet of parchment.18

With regard to the head-tefillin, [by contrast, since] they have four compartments, the initial and preferred option is that the passages be written on four different sheets of parchment, as will be explained.19

4 It is a halachah transmitted to Moshe at Mount Sinai that [the passages of] tefillin may be written only with ink.20 Moreover, there is an authority21 who rules that Torah scrolls, tefillin, and mezuzos22 may be written only with ink that can be erased, as in the passage concerning the sotah:23 “[The kohen] shall write [these curses on a scroll] and [then] blot [them] out,” [implying] that the script must be erasable. And since one traditional interpretation equates [the kohen’s] writing with the writing required for Torah scrolls, tefillin, and mezuzos, [the latter writing, too, must be erasable].24 Accordingly, vitriol25 should not be mixed with the ink, because it creates a very distinct image and cannot be entirely erased.

It is appropriate to respect this opinion as the initially preferred option and [instead] to make ink out of [soot from] the smoke of [burning] wood or from oils saturated with gallnut juice and the like, so that if one would [try to] erase it, [the writing] would be erased. Any [substance whose] smoke [produces soot] may be used for the ink,26 but [soot from the smoke of] olive oil is the most desirable.27

If, however, one wrote Torah scrolls, tefillin, or mezuzos using a mixture of gallnut juice and vitriol which is long-lasting and cannot be erased, they are valid. For the predominant view is that we do not equate the writing mentioned in the passage of the sotah to [other] writing. This is the present custom, which is followed even as an initial and preferred option.28

If so, what is implicitly excluded by the [above-mentioned] halachah transmitted to Moshe at Mount Sinai, that [the passages of tefillin] may be written only with ink? Excluded are colors other [than black], such as red, green, or the like. If even one letter in a Torah scroll, in a [passage of] tefillin or in a mezuzah was written in another color or in gold, [the article] is invalid and must be set aside.29 The script in its entirety must be black.

If gold dust was sprinkled over letters,30 the gold should be removed, and then the script remaining below is valid. If, however, one sprinkled gold dust over one of the letters in G‑d’s name, the problem cannot be corrected and [the passage] must be set aside; it is forbidden to remove the gold, for this would be like [violating the prohibition against] erasing G‑d’s name.

5 Our Sages31 interpret the word32 וכתבתם — “And you shall write them” — [as a directive to write with] כתב תם, “an unblemished and complete script.” This means that no letter should touch another; rather, each letter should be surrounded by [blank] parchment on all sides,33 and it should have the crownlets34 required by the halachah as detailed in sec. 36[:5].

If the extremity of a large letter touched [a neighboring letter] in such a manner that if the point of contact were scraped away the shape of the letter would remain acceptable, some authorities rule that [the letter] is valid even if [the point of contact] was not scraped away.35 One should, [however,] adopt the stringent view and scrape away [the point of contact]. [See fig. 37.]

In fact one should adopt the stringent view and correct by scraping even when the actual letter does not touch [another letter], but [merely] a crownlet of one letter touches the crownlet of another letter. Moreover, even if the crownlets of one letter touch each other, it is desirable to correct them, as will be explained in sec. 36[:5].

Correction is acceptable even with regard to tefillin and mezuzos,36 even in instances of touching letters which would otherwise be disqualified by all authorities, as will be explained.37

Fig. 15: When the inner edge of the letter pei follows the tradition received from R. Yehudah the Chassid, its silhouette produces the letter beis.

6 The initial and preferred option is to write with a slightly thick script, so that [the letters] will not [fade or flake off] readily. Likewise, one ought to38 beautify the appearance of the outer [edges of the] letters. How much more should one beautify their inner [edges] — just as the interior of the Beis HaMikdash was lined with gold.39

7 There are authorities who maintain that [a scribe] must write [the passages for tefillin] with his right hand, and that if he writes with his left hand, they are invalid, for this is not the common mode of writing, and indeed is [considered] anomalous. For a person who writes with his left hand on Shabbos is not [punishable for transgressing the laws of Shabbos]40 — and it is unthinkable that a person write a Torah scroll, tefillin, or a mezuzah on Shabbos and not be liable.41

Even if a person is ambidextrous, the initial and preferred option is that he write [tefillin passages] with his right hand, though if he wrote them with his left hand, they are valid. A left-handed person should, however, write with the hand that is “right for him,” i.e., his left hand.42 If he writes with the hand which is “left for him,” i.e., his right hand, [the passages] are invalid.

Nevertheless, if it is impossible to find other [passages for] tefillin which were written with the right hand, [passages] written with the left hand should be deemed valid, whether written by a left-handed person or a right-handed person,43 for this disqualification is not mentioned in the Talmud.44 If, however, a person wrote [the passages] with his mouth, i.e., he held the pen with his lips and wrote, [the passages] are invalid even if no others can be found. For certainly, writing with one’s mouth is not at all the ordinary way of writing.45

8 Unlike mezuzos, concerning which there is a halachah transmitted to Moshe at Mount Sinai that every line [of the parchment] must be scored,46 [the passages of] tefillin do not require this, except for the uppermost line.47

[This is required because,] as an initial and preferred option, it is forbidden by Rabbinic decree to write three words from a verse of the Tanach without scoring the parchment, as stated in Yoreh Deah 284:2. The other lines, however, need not be scored, for the uppermost line of text serves as scoring for the line below it,48 as explained there.

One should not adopt a stringent stance and score all the lines, because “whoever observes a practice from which he is exempt is considered a commoner.”49 Nevertheless, if [the scribe] cannot write a line straight if it is not scored, it is advisable that he score all the lines so that the script will be straight and regular, in the spirit of the verse,50 “This is my G‑d, and I will glorify Him.”

[Moreover,] even if [the scribe] is able to write [the lines] straight without scoring them, but scores them so that he will be able to write more beautifully and in straighter lines, he is not considered a commoner [for doing so]. For he is not observing this practice as a stringency, but rather in order to write straighter (and thus enhance his observance of the mitzvah). (See [the gloss of] Beis Yosef [to the Tur]).

The statement that “whoever observes a practice from which he is exempt is considered a commoner” applies only when he does so as a stringency (at a time when he will not risk a transgression if he does not follow this stringency, nor will the observance of a mitzvah be enhanced if he does follow it. If it leads to the enhancement of a mitzvah, or if he follows this practice as a protective safeguard so that he not transgress, he is not considered a commoner [for observing the stringency].

These in fact are the reasons for all the stringencies required by the later sages. They are either protective restrictions so that one will not come to transgress a prohibitive commandment, or even a minor prohibition which the Talmudic Sages deemed necessary, or they are measures for the enhancement of some positive commandment, even one which is Rabbinic in origin. If, however, [one undertakes a stringency] when the enhancement [of a positive commandment] is not involved, e.g., one sits in a sukkah while it is raining, he is considered a commoner, as stated in sec. 639[:23].51 The same [principle] applies in all similar instances.)

One should not rule the lines with lead52 or similar substances because the scored line remains tinted. After the fact, however, [this kind of scoring] is valid.

Some authorities hold that it is always necessary to score [the parchment] from above and from below53 and at the side margins, both when writing tefillin and whenever one is writing more than three words from a [Biblical] verse, even when [the scribe] is able to write [properly] without scoring lines.

The halachah rules according to the former view, as stated in Yoreh Deah, loc. cit. In the case of tefillin, nevertheless, the custom is to follow the latter, more stringent view.

9 It is a halachah transmitted to Moshe at Mount Sinai that tefillin may not be written on duchsustus or on gvil, but only on klaf,54 otherwise they are invalid. [When writing] on klaf, one may write only on the surface that faces the flesh [of the animal]; if one wrote on the surface that faces the hair, [the passages] are invalid.

What is meant by klaf and what is meant by duchsustus? In the course of being processed, the [dual thickness of a] hide, known as gvil, is split into two skins. The outer skin that faces the hair is called klaf, for it is peeled off (niklaf) from the inner skin, which is attached to the flesh. This [inner skin] is called duchsustus [lit., “at the place of the flesh”], for duch is Aramaic for “place” and sustus is Greek55 for “flesh.” The requirement that one must write on the surface of the klaf that faces the flesh refers to the surface that is closer to the flesh of the animal, i.e., the interface with the duchsustus.

[Today’s] gvil, which is not split [into two skins], is [nevertheless] considered klaf, and [the passages of] tefillin are written on the side facing the flesh. For the tough and hairy outer surface of the hide is scraped only to the extent needed to make it usable and smooth. In fact, even if the hide were to be split into two skins, it would be necessary to scrape off this much of it.56 Moreover, much is scraped off from the surface facing the flesh, until all that remains is the klaf alone.

10The klaf must be made of a hide that has been treated with gallnut juice, lime,57 or other similar substances which constrict the hide and strengthen it. Otherwise, it is called diftera. [Being incompletely processed, such parchment] is not called a scroll (ספר) in Hebrew,58 and is [therefore] invalid for the writing of a Torah scroll, for several times59 the Torah refers to itself as a scroll.

Similarly, [it is invalid] for the writing of tefillin or mezuzos. (For the fact that the term “writing” is used with regard to both a Torah scroll and a mezuzah establishes an analogy between the two.60 And a comparative association61 is established between tefillin, concerning which the Torah states,62 “And you shall bind them,” and the mezuzah, concerning which [the following verse states], “And you shall write them.” Accordingly, a Torah scroll and [the passages of] tefillin likewise require a “perfect script.”63 For with regard to a mezuzah [the Torah] states וכתבתם — “and you shall write them,” [teaching that the script must be perfect (כתיבה תמה), and because of the above associations] this requirement for a mezuzah is also applied to a Torah scroll and to tefillin.)

[When processing the hide,] one should be careful to leave it in the lime until its hair falls off unaided. If [the hide] is removed [from the lime] before that time, it may not be used for writing, for it is not fully processed and is still called diftera.

11 [The hide] must be processed with the intent that it be used for [the mitzvah for which it is intended].64 For example, [parchment to be used] for a Torah scroll must be processed with that intent in mind. [Parchment to be used] for tefillin must be processed with that intent, and [parchment to be used] for a mezuzah, with that intent.

If, however, [parchment] is processed with the intent that it be used for a Torah scroll, it is also valid for tefillin and for a mezuzah. For the sanctity of a Torah scroll is greater than that of tefillin and of a mezuzah. Thus [the parchment] is considered as if it had been processed for the sake of tefillin or a mezuzah as well, just as [the number] 200 subsumes 100.65 However, [parchment] processed with the intent that it be used for a mezuzah is invalid for tefillin, for the sanctity of tefillin is greater than that of a mezuzah.66

It is desirable to make a verbal statement [that one is processing the parchment] for the sake of a Torah scroll or for the sake of tefillin. Merely having the intent in mind is not sufficient.67

12 Some authorities disqualify parchment that was processed by a non-Jew, even if he is being supervised by a Jew68 who instructs him to process [the parchment] for the proper intent, just as a bill of divorce written by a non-Jew [for a Jewish couple] is invalid69 even if he is being supervised by a Jew who instructs him to write it specifically for this particular husband and wife.

[The rationale is] that a non-Jew acts on his own initiative and does not heed [the instructions given him by] the Jew. (Even if he says that he is taking heed, his mouth does not echo his heart, and even if he says that his mouth does follow his heart, his words are not to be believed.70)

Other authorities consider [parchment processed by a non-Jew] valid if he is being supervised by a Jew who instructs him to process it for the proper intent. This case cannot be likened to a bill of divorce, the entire core of which must be written specifically for this particular husband and wife. With regard to such [an involved task, it cannot be assumed] that the non-Jew will heed [the Jew’s instructions]. With regard to the processing of parchment, by contrast, the intent that it be used for the mitzvah is necessary only when the parchment is put into the lime. With regard to such [a specific act, it may be assumed] that [the non-Jew] will heed [the instructions].71

Nevertheless, as an initial and preferred option, [a Jew] should also help somewhat with the processing. It is customary in these countries72 [to follow this latter opinion], for Jewish parchment processors are not found in all places.

The help should be given in the actual processing [of the hide]; it is not sufficient to help blacken the straps [of the tefillin], for this is not part of the task of processing [the parchment].

After the fact, even if [a Jew] did not help [the non-Jew with the processing of the parchment] at all, but merely directed him to put it in the lime for the sake of the mitzvah, [the parchment] is valid. If a Jew puts it in the lime for the sake of the mitzvah, he may allow the non-Jew to remove it and prepare it alone; there is no need to help him further.

13 If hides given to a non-Jew to process for use [in the mitzvah for which it is intended] are identified by signs resembling letters, which the owner punctures with an awl so that the non-Jew not exchange [these hides] for others, [we do not suspect that the non-Jew will forge this sign]. Although this is easy, for he could place the [punctured] hide on top of another hide which he could now puncture [identically] by inserting an awl in the holes made by the Jew, we do not suspect that he will do this — either because he will be afraid the Jew will recognize his own signs, or because the new holes will appear more recent than those made by the Jew.

[The question of whether] a hide which was processed without the intent of being used for a mitzvah can be corrected by reprocessing it for the sake of the mitzvah is discussed in Yoreh Deah, sec. 271.73

14 The parchment must come from the hide of a domesticated animal, wild beast, or fowl, of a kosher74 species, even if that creature died without ritual slaughter75 or was deemed a treifah.76

It may not come from the hide of a domesticated animal, wild beast, or fowl, of a non-kosher77 species. [This restriction is derived78 from] the verse,79 “[And it shall be for you as a sign upon your arm and as a reminder between your eyes,] so that the Torah of G‑d shall be in your mouth.....” [The connection implied by the verse teaches that the tefillin may be made only from species] that are permitted “in your mouth.”

Nor may [the parchment for tefillin be made from] the skin of a fish, even if it is of a kosher species, because [such parchment] is malodorous.

Similarly, whenever G‑d’s name is written, even when it is not [written] for a Torah scroll or tefillin or mezuzos, it may not be written on parchment from a non-kosher species of animal or the like.

15 Conscientious scribes make three types of parchment: a thicker one on which to write the passage beginning Shema, which is short, a thinner one for the passage Vehayah ki yeviacha, which is longer, and for the passages Kadesh and Vehayah im shamoa, which are long, they prepare very thin parchment. All the [four] compartments [of the head-tefillin] are thus filled [with scrolls of] equal [diameter].80 This beautifies the tefillin.

16 As explained above,81 the script must be perfect and complete. There must therefore be no holes in the parchment in the places where one is writing. If one wrote in a place where a hole is located, [the passage] is invalid, because such script is not perfect and complete, but interrupted.

Even if the hole does not intercept the entire width of [one of the strokes of] the letter from one side to the other, but is located in its midst and is surrounded by ink on every side [as in fig. 16], [the letter is still invalidated,] for where the hole is located, the letter is intercepted and divided in two. [This applies] even if the width of the script on either side of the hole [still] meets the [minimum requirements of] thickness required for this letter [on either side of the hole].82 [See fig. 17.]

Moreover, the letter [is invalid] even if the length of the letter until the hole also exceeds the minimum required length of this letter [see fig. 18], for the hole is considered to be intercepting the body of the letter. The long part of the letter that reaches up to the hole is thus not surrounded by [blank] parchment, nor is the width of [the stroke of] the letter [that remains] at the sides of the hole surrounded by [blank] parchment — and the entire letter must be surrounded by [blank] parchment on all sides; no part of it may be left without [blank] parchment surrounding it.83

[In the above case of a hole located in the midst of the width of the script,] the [blank] parchment at the outer edge of the stroke on one side of the hole cannot serve [as a surround] for the stroke [that borders the hole] on the other side, because the hole intercepts them and divides them in two. In the same way, the [blank] parchment at the end of the [shorter] part of the letter beyond the hole cannot serve [as a surround] for [the longer part of] the letter that reaches as far as the hole.

(Nevertheless, [since the script bordering the hole meets the minimum requirements for the width of the letter], one may scrape away [the ink] at the side of the hole, so that there will be blank parchment between the hole and the width of the letter at its side. Similarly, if it is possible to scrape away part of the [length of the bulk of the] letter above the hole, and yet leave the minimum length of the letter, this [adjustment] is sufficient [to render the letter valid].

This applies even if one has already written many [letters] after this letter. One is not considered [by the subsequent adjustment] to have written the letters in incorrect sequence, for there is nothing [intrinsically] invalid about [the shape of] this letter; it is only that it is not surrounded by [blank] parchment, as will be explained.)84

17 The above applies {with regard to a hole that was made before the writing}85 when the empty space of the hole can be seen by the eye86 as dividing the letter in two. Whenever, by contrast, ink passes over a hole and closes it, and it does not appear as a hole except when held up to the sun, it is not considered a hole at all. One may write on [such parchment even] as an initial option.

For this reason, license was granted to write on parchment made from the processed skin of a fowl, even though it is entirely filled with small holes where the feathers were located.

18 The above applies with regard to a hole that was made before the writing took place. Thus, at the time of writing, the script was already intercepted by the hole and was not perfect and complete.

[Different rules apply, however,] if the writing was perfect as required by law [at the outset], but was perforated later. Even if the hole intercepts the entire width of [one of the strokes of] the letter from one side to the other and [the letter] appears to be divided in two [see fig. 19], it is valid if the length [of the remaining bulk of the letter] that reaches as far as the hole meets the minimum requirements for that letter without any need to consider the part [beyond] the hole.

Moreover, it is valid even if the part until the hole does not meet the minimum requirements for [the length of] the letter, but the hole does not intercept the entire width of [the stroke of] the letter [see fig. 20], and a thin, uninterrupted strip of its width remains at the side of the hole. For if its width had been made this thin at the outset, it would [also] have been valid, since [the Sages] did not prescribe any minimum width for the [component strokes of the] letters; they may be made as thin as one desires.

19 What is the minimum acceptable length for a letter that is intercepted by a hole? If the right leg of a hei, or of a similar letter, such as resh or dalet, has been intercepted [see fig. 21], and what remains above the hole is as big as a small letter, i.e., a yud, [the foreshortened letter] is valid.87

If it is the left leg of the hei that is intercepted by a hole, some authorities rule that the letter is valid even if only a minute part of the leg remains, whether above the hole or below it. Others maintain that the left leg must also be the size of a yud [see fig. 22], and that otherwise the letter is invalid [see fig. 23]. The halachah follows the latter opinion.

If, however, the central space of a hei, a mem, or the like, was perforated [see fig. 24], then even if the hole occupies all of that space so that the letter is not bordered [along its inner edge] by [blank] parchment, it is valid. Similarly, if the outer edge of a letter was not surrounded by [blank] parchment because [part of] that edge was perforated by an elongated hole extending to the edge of the parchment [see fig. 25], the letter is nevertheless valid,88 since it was surrounded by [blank] parchment at the outset, when it was written.

By contrast, [a letter] that was not surrounded by [blank] parchment at the outset is invalid, even if it has the required length and is not severed by a hole at all. In the last line of a column, for example, if the long leg of a final chaf or nun or similar letter reached the lower edge of the parchment from the outset [see fig. 26], the letter is invalid. Nevertheless, if it is possible to scrape away a small part of the letter at the edge of the parchment without encroaching on the letter’s minimum length, this procedure is effective even if the entire passage has been completed. [The passage is] not [disqualified] thereby on the grounds that it was not written in the proper order, for the letter was never inherently invalid; [the difficulty was] only that it was not surrounded by [blank] parchment, as will be explained.89

Other authorities maintain that if a perforation occupies the entire central space of a hei or a mem or a similar letter [see fig. 26], such a letter is invalid because it is not bordered along its inner edge by [blank] parchment, even though it was bordered in this way at the outset.

[Suppose,] however, the perforation does not occupy the whole of that central space, but adjoins [part of the inner edge of] the letter so closely that there is no [blank] parchment whatsoever between [that side of] the letter and the hole within it [see fig. 27]. In such a case, even though a letter must be bordered by [blank] parchment on all sides, [the above-described letter] is considered to be bordered by [blank] parchment even at the site of the hole, by virtue of the [blank] parchment within its central space, on the other side of the hole.

Similarly, if a hole indents the outer edge of a letter, leaving the inner part of the letter intact [see fig. 28], the letter is valid by virtue of that inner part. That part is considered to be surrounded by [blank] parchment, by virtue of the [blank] parchment on the other side of the hole.90

{Similarly,91 if [as in fig. 19 on p. 256] a hole intercepts the entire width [of the script], but [the part of the letter] that remains until the hole has the minimum required length, the letter is rendered valid by this minimum length, which is considered to be surrounded by [blank] parchment even at the lower [i.e., further] extremity of the letter by virtue of the blank parchment below [i.e., beyond] the hole. (If part of the length of this [letter] remains below the hole, this does not disqualify the letter, and it is rendered valid by virtue of the blank parchment at the lower extremity [of the severed remnant] of the letter. For since [this severed remnant] extending from the hole until the [blank] parchment does not constitute another letter, it is not considered as separating the letter proper from the [blank] parchment which surrounds it, just as the hole is not considered as a separation.) If, however, no blank parchment whatsoever remains below the hole, [the letter] is invalid.}

According to this line of thinking, it makes no difference whether the hole existed before [the letter] was written, or appeared afterwards.92 For the blank parchment beyond the hole is considered as adjoining the letter, enabling it to be considered being surrounded by [blank] parchment [despite] the presence of the hole. Accordingly, even if the hole existed before the time of writing, it does not disqualify [the letter] any more than [if the hole appeared] after the time of writing. ([The rationale is that] part of the letter, having the minimum required size, remains intact and unperforated).93

The halachah rules according to the first approach. Nevertheless, one should respect the latter approach and rule stringently when there is no blank parchment at all on the other side of the hole, whether the hole was outside the letter and immediately adjacent to it, or within the letter and occupying its central space entirely.94

Some authorities maintain that even according to the second opinion, even after a letter was written it must be bordered by [blank] parchment only along its outer edge, but not along its inner edge. [Thus,] even if a hole occupies the entire central space of a letter, the letter is valid.95 It is, however, proper to follow the more stringent ruling [and not accept this leniency].

20 If one of the straight letters that have only one leg, such as a vav or a zayin or a final nun, is severed, the fact that the remnant is as big as a small letter, i.e., a yud, does not [necessarily] render it valid, as is true with regard to the letter hei and the like.96 For when all that remains of a vav is a part the size of a yud, it is no longer a vav. The same applies with regard to the other straight letters.97

Instead, the criterion for their validity is how they are read by a young child who is neither bright nor slow. If such a child can read the letters properly, they are valid; if not, they are invalid.98

We cannot rely on a bright child who understands the meaning of the words, because his reading of the severed letter is guided by how it will fit the context. If, however, the child is not bright enough to understand the meaning but is bright enough to recognize the letters well, he may be relied upon. This is the intent of the Talmudic phrase, “nor slow.”

From this explanation you may deduce that there is no need to cover the letters [adjacent to the letter in question], as is customary. For were this to be done, how could even a bright child who understands the meaning of the words know how to read this letter correctly? There would be no way for him to know its context and meaning.

Nevertheless, if a letter was severed by a hole and part of it remains below the hole, the remnant below the hole must be covered. For [in his mind] the child will join it with the [main] part above the hole and read the letter properly. And in truth, since the hole separates [the two parts of the letter], they are not considered as being joined together.

The above laws also apply if a scratch crosses the width of [one of the strokes of] a letter from one side to the other. If, however, the scratch does not pass from side to side, even if it existed at the time of writing, it does not disqualify [the letter], for it is not completely divided in two. A distinction must be drawn between [such a scratch] and [the above case99 of] a hole located in the midst of the width [of one of the strokes of a letter]. If such a hole existed at the time of writing, it disqualifies the letter, for the part of the letter’s width [that remains] at either side of the hole is not surrounded by [blank] parchment. In the present case, by contrast, the scratch is blank parchment. Thus it allows the part of the letter’s width that renders the letter valid to be considered as being “surrounded by [blank] parchment.” For [the Sages] did not prescribe any minimum width for the [component strokes of the] letters, and they may be made very thin, even as an initial and preferred option.100

[Different rules apply] if the scratch crossed [the width of one of the strokes of a letter] from one side to the other, and existed at the time of writing. Some authorities maintain that the validity of such a letter cannot be determined by a young child’s reading, even if the lower part is covered [see fig. 29], for [the Sages] applied this criterion only with regard to a letter that was written with the required length at the outset, and only later was shortened by a hole or a scratch.

Other authorities maintain that if a young child reads a letter correctly, it is valid even if it was short at the outset. Therefore, [we may rely on a young child’s reading] if a scribe writes one of the straight letters, [continues to write other letters,] and then is unsure whether the length [of the straight letter] meets the minimum measure required by law. [If it was not long enough,] he would not be able to correct [the letter], because then the letters would not be written in the correct order, as will be explained.101 [In such a case, the scribe] may show [the letter] to a young child who is neither bright nor slow and may rely on him. (The halachah rules according to the latter approach and it is the one that is commonly followed.)

21 A letter may be deemed valid when read by a young child only when there is a doubt whether the length of a letter meets the minimum requirements or whether the width of a letter meets the minimum requirements. For example, if a dalet was written small, like a yud, and a young child reads it as a dalet, it is valid.

If, however, we know that a letter was written incorrectly, then even if a young child reads it correctly this is of no consequence. If, for example, some of the yud-[shaped heads] of the letters ayin and shin and alef,102 or the legs of the letter taf,103 are not touching the bodies of their respective letters, such letters are invalid even if a young child reads them correctly. For of what consequence is the child’s reading when we see with our own eyes that the letter was not written validly?

Nevertheless, [the fact that] the child reads these letters [correctly] can be of significance, for the scribe may correct them even after having written [other letters] afterwards. This will not be considered as if the letters were written in incorrect sequence, as will be explained.104

22 [The following rules apply] if a drop of ink fell in the center of a letter, so that the letter is no longer recognizable as such (even though its inner edge remains surrounded by [blank] parchment); for example, [a drop of ink] fell in the center of a dalet and made it appear to be a hei [see fig. 31], or into a beis and made it appear to be a pei [see fig. 32], or made it no longer be deemed a beis. Each of these letters, and [likewise] any other letter, is invalid [if a drop of ink] falls in the center of it, changing its appearance to the extent that it would no longer be referred to by its name,105 or if the ink detracts somewhat from [the shape of] the letter — even if it merely detracts from a point of the letter, when the lack of that point disqualifies the letter.106 [Similarly,] if a drop of ink appears as [part of] the body of a letter, then even if the letter [remains] clearly recognizable as such, it is invalid.

The problem cannot be corrected by scraping away the [dried] drop of ink so that the letter thereby acquires its correct shape, for this would be considered chak tochos,107 i.e., engraving the interior of the letter so that it is left with its statutory shape by virtue of the ink that was not scraped away. And [with regard to mezuzos — and, by association, this also applies to tefillin] the Torah states, “You shall write them,” not “You shall engrave them.” Our Sages108 interpret this phrase as referring to [and excluding] chak tochos.

The same laws apply if one erases ink while it is still wet, and in this way the letter acquires its correct shape. This [too] is considered as chak tochos, for there is no difference between engraving, scraping, and erasing. None [of these] constitutes writing — and the Torah states, “You shall write them.”

23 Similarly, if one erred and wrote a dalet instead of a resh, or a beis instead of a chaf, [the letter] cannot be corrected by scraping away or erasing the protrusion, because this is considered chak tochos. One may, however, add ink to make the letter [thicker and] rounded like a resh or a chaf.

This may be done only if one has not yet written anything else after this letter. Otherwise, [the passage] is invalid because it was not written in the proper order, as will be explained.109

Similarly, if one wrote a resh instead of a dalet, or a chaf instead of a beis, [the letter] may [be corrected by] adding ink and making [the corner] square as [required by] law, provided one has not written [another letter] afterwards. For whenever a correction is made by writing and not by scraping away or erasing, it is valid. One may not, however, scrape away [part of] the chaf so that it will appear as a beis, nor [scrape away part of] a resh to make it appear as a dalet.

24 The above applies to scraping away ink. If, however, wax falls on the letters, one may scrape it away, and [the letter is] valid. The question of whether it is permitted to scrape wax from a Torah scroll on Shabbos will be explained in sec. 340[:4].

25 If the gap in an open mem110 becomes closed, it may not be corrected by [merely] scraping away the connecting [ink] and thereby opening the gap, for this is considered as chak tochos.

How should it be corrected? [The scribe] should scrape away the entire “nose,” [i.e., the left side of the letter, as in fig. 34], leaving [what resembles] a curved nun,111 and then he should [re]write what he had scraped away. This is not considered chak tochos, even though he did not scrape away and rewrite the nun [as well]. [The rationale is] that the shape of the nun was originally written validly. It was merely that the letter became disqualified when the “nose” touched it. It is therefore sufficient to scrape away the “nose” and write it anew.

If, however, a resh is written so that it resembles a dalet, one should be stringent and [follow the view that] maintains that it is not sufficient to scrape away the leg alone or the upper line112 alone, and rewrite it. [The rationale is that since the problem is in the corner,] both the upper line and the leg were written invalidly. For it is the practice of scribes to first write the upper line [of a dalet] together with the protrusion at its [right] extremity,113 and then [to add] the foot below it. Thus the upper line was also written invalidly, because [the scribe] added the protrusion at its [right] end.114 Needless to say, this applies to the leg, which was written after the protrusion was formed.115 Both [parts of the letter] must therefore be scraped away.

It could be argued that when one scrapes away either of [the two parts], what remains does not have the shape of a letter at all; hence, when one rewrites the part that was scraped away, this is as if one were writing the entire letter anew. Nevertheless, one should not follow the lenient view and leave any vestige of the letter at all, for the whole letter was formed invalidly. True, the upper line was written validly and it is only the squared corner behind it that is invalid, but since it was all written in one stroke, it is considered as a single unit [and disqualified in its entirety].

A similar case: The leg of a final chaf116 must be twice as long as its upper line, so that it not resemble a resh. If [a scribe] made its upper line as long as its leg, then if he cannot extend the leg to twice the length of the upper line,117 the upper line must be completely erased because it was entirely written in one stroke. One may [correct the letter by] erasing [this part], even in the word Elokecha,118 provided one is careful not to erase the leg.

This does not apply to the “nose” of the mem, for it is not written in one stroke together with the part [to the right] that resembles a nun. Similarly, any letter that is written in two strokes, such as a gimmel and the like, is governed by the same laws as a mem.

26 Nevertheless, any part of such a letter that was written after it was disqualified must be scraped away and written anew, even though it was not written in the same stroke as the disqualifying factor.

To cite an example: A yud at the upper [right] of the alef which touches the [diagonal] shaft below it more than necessary [see fig. 35]. As explained in sec. 36, it is only the protrusion [that descends from] the right [of the upper yud] — and not the body [of the yud] — that should touch [the shaft]. If a scribe wrote the shaft of the alef so close [to the upper yud] that it touches the body of the yud [and then wrote the lower yud], he must scrape away and rewrite the entire shaft and the yud below it.

If, however, the point at the [lower] left [of the upper yud], but not the body of the yud itself, touches the shaft [of the alef], the letter is valid. Even if that point were lacking, this letter would still be called a yud; therefore, the fact that it is touching [the shaft] does not disqualify [the letter].

If the body of the lower yud [of the alef] is touching the shaft there is no need to scrape away the shaft, but only the lower yud itself, for this is what was written invalidly.119 The same principle applies with regard to the [left] leg of the letter taf.

27 Similar laws apply if any of the yud-[shaped components] of a shin, an ayin, a tzaddik120 or a pei touch the actual letter with their bodies rather than with their protrusions. One must scrape away every part of the letter that was written after the disqualifying factor came into being. The fact that a young child reads the letter correctly is of no avail since we can see for ourselves that the letter is written incorrectly.121

Similarly, if the [left] leg of a hei or a kuf touches the upper line of its respective letter, even as slightly as a hairsbreadth, the letter is invalid, even if a young child who is not bright reads it correctly. [See fig. 36.] Moreover, [these letters] cannot be corrected by separating [the parts that touch], because this is considered as chak tochos. Nevertheless, there is no need to scrape away the entire letter; it is sufficient to scrape away the leg that was written invalidly and write it afresh. The upper line and the right leg that were written validly at the outset do not have to be scraped away.

If, however, one initially wrote a resh, and then erred and made a protrusion [at its upper right extremity] so that it appeared like a dalet, one cannot correct it merely by scraping away the protrusion. By doing so one is not working on the letter itself; rather, the letter is transformed into a resh merely by his scraping; hence this is considered as chak tochos. This is not comparable to [the above-mentioned correction of] the hei and the kuf, where action is taken on the letter itself. In those cases, since one scrapes away the leg and writes it afresh, this is considered as if he had written the entire letter, for without this leg the letter would not be called a hei or a kuf.

Based on the above, there is also room for [some] leniency [in the instance described above, in which] one wrote a resh [correctly] at the outset,122 but erred and added a protrusion that made it appear as a dalet. [It is sufficient] to scrape away the protrusion as well as the upper line alone, or [the protrusion] as well as the leg alone. In this way, one will be taking action on the body of the letter. There is no need to scrape away the remainder [of the letter], since it was originally written validly. Nevertheless, if a [scribe] is stringent and scrapes away the entire letter, may blessings light upon him.

28 The above applies when the disqualifying factor involves the shape of the letter itself. The disqualification of chak tochos does not, however, apply when letters are touching each other and attached together. Leniency is granted whether the letters became attached before the second letter was written completely and thus never acquired the shape of a valid letter or whether they became attached after [the scribe] finished writing it. One may separate them [by scraping away the point of connection] with a knife and then they are valid. [See fig. 37.] [The rationale is that] the disqualifying factor did not involve [the shape of] the letters themselves, but rather the fact that they are not surrounded by [blank] parchment.

Accordingly, if a letter had its shape changed by becoming attached to another letter, separation is not effective and it is governed by the laws applying to a mem which became closed;123 i.e., whatever part was written invalidly must be scraped away.124

Some authorities maintain that [stringency is also required when] the entire length of one letter is attached to another one; here, too, [merely] separating the letters is insufficient.

29 Whenever [merely] separating the letters is not effective, overwriting a letter125 is [also] not effective; it does not remove the need to scrape away [the problematic letter] and rewrite it. As long as the initial script exists, any overwriting is of no consequence, for writing upon existing script is not considered writing at all. Only [writing] on blank parchment [is considered writing].

30 Whenever [merely] separating the letters is not effective and one must scrape away the letter and rewrite it, or even a small part of it, this correction is of value only when one has not yet written anything after the error.

If, however, one wrote [other letters], he must erase everything he wrote afterwards and rewrite it all after correcting the error. If [after the error] he wrote one of [the Divine] names which may not be erased, this passage [cannot be corrected and] must be put aside.126

The reason: When one corrects the error by rewriting [the letter] after having written a later letter from the passage, the passage will not have been written in the proper sequence. [This is unacceptable] because the Torah states,127 “And these [words] shall be....” Implied is that they shall come into being [in the order which they appear in the Torah].128

When, however, letters are attached to each other, so that they may be corrected by separation, one may separate them even after the entire passage has been written [see fig. 37]. One is not considered thereby to be writing [the letters] in incorrect sequence, since one is not writing them at this time, but rather correcting them by scraping away [the ink] that attaches them. And scraping is not writing.

[This law can be extended and applied] even when a slight amount of writing is necessary to correct a letter. If its correct shape was already defined when it was [first] written, so that it would not at all be mistaken for another letter — if, for example, [one of] the yud-[shaped components] of a shin or an ayin or a tzaddik or an alef or a pei, or if the leg of the letter taf, was not touching the body of its respective letter, and a young child who is neither bright nor slow would read the letter properly — it may be corrected afterwards. [See fig. 30] This is not considered as writing the letters in incorrect sequence.

If, however, the [left] leg of a hei or a kuf touches the upper line of the letter even as slightly as a hairsbreadth [see fig. 36], [the above does not apply, even though] the correct shape of the letters was already established and a young child will read them properly. Nevertheless, since the entire leg must first be scraped away [to correct the letter], the letters will no longer have their correct shape, and when one [retraces his steps and] rewrites them, he is deemed to be writing in incorrect sequence.

Similarly, whenever other [parts of] a letter are touching in a manner in which [some parts of] the letter must be scraped away to the extent that it loses its required shape, it cannot be corrected afterwards. To cite examples: the upper or lower yud of an alef, or [the yud] of a shin or of an ayin or of a tzaddik or of a pei, or the leg of a taf, that are touching the body of their respective letters unduly, to the extent that the letters are invalidated. [See fig. 35] This law also applies to an open mem which became closed129 and other similar situations.

Moreover, even when there is no need to scrape anything away at all, but merely to [add] a minimal amount of writing, nevertheless, if a letter is divided in two and appears to be two letters — e.g., a tzaddik which is written as a yud and a [separate] nun [see fig. 38], or a shin which is written as an ayin and a [separate] yud — then even though a young child who is not bright will read it properly, it cannot be corrected afterwards. For we see with our own eyes that such letters have the shape of other letters. Thus when one corrects these letters after [having continued to write], it is as if one is writing anew, and [the passages are considered] as not having been written in the correct sequence.130

Likewise, when a ches is written as two zayinim,131 and as soon as one sees them it is clearly apparent that they are separate, [correcting the letter after having written further is considered as writing in incorrect sequence]. If, however, the separation at the “hump” above the zayinim is not apparent unless one looks at it [closely], the letters can be corrected afterwards. [This applies] even if a child reads them as [two] zayinim. With regard to this letter, we do not follow [the reading of a child], because a young child is not used to seeing a ches [written] in this manner with a “hump.”

(With regard to a tzaddik or a shin, by contrast, we follow the child’s reading, whether it results in a lenient ruling or a stringent one,132 unless we clearly see for ourselves that [these letters] are written as a yud and a [separate] nun, or an ayin and a [separate] yud. In such an instance, the child’s words are of no consequence.)

Similarly, if [a scribe] wrote a yud without a right leg, the letter may not be corrected after [having continued writing], even if a young child reads it properly, because without this leg it cannot be called a yud, and thus one would be writing in incorrect sequence. If, however, one did not make a protrusion on the [lower] left side of a yud, this can be corrected later, because even without this protrusion the letter is called a yud. [Adding this protrusion] is thus not considered as writing in incorrect sequence.

31 [The passages] must be written for the sake of the mitzvah,133 i.e., one must write them with the explicit intent that they be used for the sacred purpose of tefillin. If one writes them without [this] specific intent, they are invalid.

[Moreover,] each time G‑d’s name appears,134 it must be written explicitly for the sake of its holiness.135 If it is written without [this] specific intent, [the passages] must be put aside.

There is an unresolved question [in Jewish Law]: Whenever an activity must be undertaken for the sake of a particular mitzvah, is it sufficient that one have the intent in one’s heart that he is doing it for its designated purpose, e.g., writing the passages of tefillin for the sake of their holy purpose and writing G‑d’s name with its sacred intent, or must one verbalize that intent, thought alone being insufficient?

Since the unresolved doubt involves a point of Scriptural Law, one must follow the more stringent opinion. [Therefore,] when a person begins to write, he must state, “I am writing these passages for the sake of the holiness of tefillin.” Moreover, whenever he writes G‑d’s name, he must state that he is writing it specifically for the sake of the holiness of G‑d’s name.

If [the scribe] did not state this explicitly when writing G‑d’s name, but had the intent in his heart, what he has written is acceptable after the fact, since he did make an explicit statement when he began to write that he was writing [these passages] for the sake of the mitzvah.136

If, however, he did not verbalize this intent when he began to write, one should be stringent — because a point of Scriptural Law is involved — and disqualify [the passages] even after the fact, even though [the scribe] had [the proper] intent in his heart.

With regard to the processing of hides [for parchment], by contrast, one may be [more] lenient. [Parchment is acceptable] after the fact if, at the beginning of the process, the [craftsman] had the intent in his heart that he was processing the hides for use in a Torah scroll or for tefillin, even though he did not state this explicitly.

[The rationale is that] some authorities do not require that the parchment be processed with the intent that it be used for a sacred purpose. [They maintain that] doing so is merely the choicest way of performing the mitzvah, but after the fact, [the parchment] is acceptable even when processed without this intent.

Even though the halachah does not rule according to this view, one may rely on it to reach a lenient ruling after the fact when [the craftsman] has had the intent in his heart without verbalizing it — for perhaps thought alone suffices, even initially.

32 When [a scribe] is dozing off he should not write [tefillin], for then he is not writing with the intent of the mitzvah. (For although a verbal statement is necessary only when one begins to write, one must have the intent in mind that he is writing for the sake of tefillin throughout the entire [time he is] writing. One may not write as a casual activity. To cite a parallel: While writing the operative clauses of a bill of divorce, [the scribe] must have in mind throughout that he is writing for the sake of this particular husband and wife, as explained above.)137

33 [A scribe] must be very careful [to write the words in the tefillin as they are written in the Torah], being precise as to which words are to be spelled in brief and which are to be spelled in full,138 for if one subtracts or adds a single letter [from or to the traditional text], the tefillin are invalid. People who put them on will thus be reciting a blessing in vain every day, and also will be lacking [the observance of] the mitzvah of tefillin every day. The punishment of the scribe will thus be weighty.

For this reason, those who engage in the writing and preparing of tefillin must be permeated by the fear of G‑d and stand in great awe of His word.

[Accordingly,] one must warn and protest against those scribes who allow their youthful apprentices to write [passages for] tefillin or mezuzos, and then, if they see that they are written correctly, they sell them. [By doing so,] they cause people at large to sin, and ensnare them with the use of tefillin and mezuzos that are invalid. For the youth will act immaturely, and when writing them will only think about training himself and practicing an attractive handwriting. His intent will not be directed towards any holiness or mitzvah whatever.139

The retribution [to be visited upon] the scribe will be great; he will be made accountable and will be severely punished. Of him the verse says,140 “Cursed be he who performs the work of G‑d deceitfully.” Every scribe should therefore take care regarding this manner, for his own benefit.

34 It is advisable to experiment with the quill141 before beginning to write a passage, so that it not have excess ink that might spoil [the script]. Similarly, one should be careful [and check] that the quill is fit to write G‑d’s name. [I.e.,] one should not begin by writing G‑d’s name directly [after dipping the pen in the inkwell], so that [the letters of] His name will not be marred by excessive ink. Instead, one should follow the practice which will be described in Yoreh Deah, sec. 276.142

Similarly, whenever [a scribe] writes G‑d’s name, he should first make a point of checking and reading through everything written previously in that passage, making certain that there is nothing there to disqualify it,143 so that the Divine name will not be have to be put aside on his account.

After completing a passage, one should read it in its entirety with concentration and careful attention two or three times before one begins to write the second passage. This should also be done after completing the second and the third passages.

In addition, it is desirable that one reread every passage of the head-tefillin before placing it in its compartment, lest it be exchanged for another [passage].144

Why must one go back and carefully check over every passage after one completes it? Because if one finds a letter missing in one passage, not only is that passage invalidated, but everything [written] afterwards is also invalidated thereby, for after [letters] have been rewritten in the early passage, [the later passages] will have been written in incorrect sequence.

For this reason one may not [insert] the missing letter by suspending it between the lines, as is done with regard to a Torah scroll.145 For with regard to tefillin it is written, “And these [words] shall be....” Implied is that they shall come into being [in the order in which they appear in the Torah]. The same applies with regard to a mezuzah.

35 If, by contrast, one writes a superfluous letter, this can be corrected by scraping it away if it is at the end of a word or at its beginning. If, however, it is in the middle of a word, correction is sometimes not effective, for when [the superfluous letter] is scraped away, [the word] appears to be two words.

Sometimes, however, correction is effective — when one scrapes away the superfluous letter and extends the preceding letter to fill the vacated space, so that the word will not [remain] divided. For example, if one wrote the word לאבותיך in its full spelling, including a vav after the beis, one may scrape away [the vav] and extend the preceding beis to fill its space.146 Similarly, if a correction is dependent on a [preceding] chaf, dalet, or resh, it is possible to extend them somewhat to fill the space of the superfluous letter which was scraped away.

If, however, [one of] the letters which can be extended follows the superfluous letter that would have to be scraped away, it is impossible to extend [it] backwards147 without first scraping away part of [it]. For example, if one wrote the word שאר in its full spelling, [mistakenly] interposing a vav after the alef, it is impossible to extend the resh backwards unless one first scrapes away its leg. By doing so, [however,] one deprives it of its shape. Thus when one [goes back to] correct it, one will be writing in incorrect sequence.

There is no remedy unless one can fill the blank space by slightly thickening the preceding and following letters. This is not considered to be changing the shape of the letter. When, by contrast, the letters which can be extended precede the superfluous letter, one can extend them considerably without detracting from their [standard form].

36 There is, nevertheless, a question in [the following situation]. The word מצת should be spelled in brief, [i.e., without a vav]. If one wrote it in its full spelling, [adding a vav, there is a question] as to whether lengthening the base of the tzaddik is sufficient after scraping away the vav. It could be argued that as long as one does not extend the upper part of the letter, the word is still considered as [divided in] two.148

Similarly, the word נתן should be written [thus,] in brief. There is a question whether, [if it was written in its full spelling,] it is possible to validate the word by elongating [only] the base line of the [preceding] nun after erasing the vav. At any rate, whenever a word appears to be two words, it is invalid.

If, however, when [a scribe] began writing, he [first] extended the base line of a nun or a tzaddik [to the left] and then wrote the adjoining letter above that line149 — like this:

If [such a letter] is extended exceedingly to fill in a line, its validity is questionable.151 One should [therefore] show it to a young child who is neither bright nor slow, as is done with regard to any letter of doubtful [shape], as will be explained in sec. 36[:4].152

37 If [a scribe] wrote an [entire] superfluous word, he should scrape it away and leave its place blank. For though there should not be blank space [in a passage of tefillin, after the fact] this does not disqualify [the passage], provided that the space [between the words] is not as great as the space [which must be left] between two passages, i.e., the width of nine letters.

If the word preceding the space ends in a hei or in a kuf, one may not extend these letters to fill in the space, because the [left] leg of the hei or the kuf must be positioned at the end of the letter. And if one would scrape away the leg and write it [anew] at the [elongated] end [of the letter], one would be writing in incorrect sequence.153

The [following] leniency may, however, be granted if, [in the midst of] a passage, a space the width of nine letters will be left [between words] when a superfluous word is scraped away. This would leave a space equivalent to the space [that ordinarily indicates a break] between one passage and another, and [this misplaced space] would disqualify the tefillin. In such a case, one may extend [the upper line of] a hei [to the left] even though the leg will no longer be at the [left] end of the letter. The rationale [for ruling leniently] is that this [disqualifying factor] is not mentioned in the Talmud.154

38 If [a scribe] was writing and erred, and erased [the script] while it was still wet, or [let] it dry and scraped it away — and similarly, if ink fell on parchment and he erased it while it was still wet, or scraped it away after it dried — it is permissible to write even the Divine name on the spot that was erased or scraped, as is permissible with regard to a Torah scroll.

Initially, however, it is preferable not to erase [the ink] while it is wet, but rather to dry it well near a flame, for then it will be scraped away easily without leaving any traces.

39 If one of the letters of the Divine name becomes attached to another, one may separate them by scraping away [the point of contact]. This is not considered to be erasing the Divine name; [on the contrary,] it is thereby being corrected.

40 [The following laws apply when] letters or words are slightly [flaked or faded]. If the impression is recognizable to the extent that a young child who is neither bright nor slow can read it, it is permissible to trace over [the letter] with a quill to improve the writing and renew it. This is not considered as writing in incorrect sequence, since even at this time, [before being restored,] the script is valid. By adding to it, [the scribe] is only ensuring that it not fade more.

When does the above apply? When some of the color of the [actual] ink remains. If, however, all the [actual] ink has flaked off the parchment and only a reddish trace of the disintegrated ink remains, [a scribe] who now traces over it with a quill is considered to be writing in incorrect sequence.

Needless to say, if at the outset [some of the letters] were written with ink that was not black, but dark off-white or reddish, making it necessary to trace over them [with black ink], [the passage is invalid,] for these letters are now being written in incorrect sequence.

41 Initially, one should take care that the head of a lamed from one line not intrude into the gap of a hei or a ches in the line above it, even if it does not touch [the letter].155

If [the head of a lamed] intrudes into the gap of a letter in a way that deprives the letter of its due shape, e.g., it enters the gap of a dalet and makes it appear to be a hei, the letter is invalid.

42 [A scribe] who is not thoroughly fluent in the passages must copy them from a text, lest he err.156 If he is thoroughly fluent in part of a passage, he may write that part from memory but must copy the remainder from a text.

43 When [a scribe] writes [passages] by heart because he is thoroughly fluent in them, he must articulate each word before he writes it, lest he err and write later words first, because he is familiar with them. Even if another person dictates [the passages] to him word by word, he must also articulate each word [before he writes it].

If, however, [such a scribe] is writing from a text, he does not have to pronounce every word, for seeing it written before him is as effective as if he were articulating it.

When does the above apply? When he knows the passages thoroughly, but nevertheless copies from a pre-existing text. If, by contrast, [a scribe] does not know [the passages] thoroughly and must write from a pre-existing text, seeing the passages in writing is not sufficient to free him from the requirement to articulate each word before he writes it.

44 Only a person who can read is permitted to write, even [if he copies] from a text. For a person who does not know how to read, even if he knows how to [copy and] write the letters, can easily err without realizing it.

45 [When writing,] one should leave a blank margin of approximately [the length of] half a fingernail157 above and below the ruled lines of each passage. Some authorities maintain that the height [of the margin] above [each passage is designed to accommodate] the head of a lamed, and below the passage, to accommodate [the foot of] a final chaf or nun, so that they too be surrounded by [blank] parchment if they are to be [written] there.

Other authorities maintain that the upper blank margin of approximately half a fingernail should [begin] above the head of a lamed, and that the lower one should [begin] below the [foot of] a final chaf or nun. Initially, one should respect this opinion.

There is no obligation to leave [a blank margin] at the beginning and end of a passage.158 The scribes, however, have adopted the custom of leaving a small margin at beginning and end.

46 The initial and preferred option is to be careful to leave a space the [width] of a small letter, i.e., of a yud, between words. After the fact, however, one should not disqualify [a passage] unless [the two words] appear as one word to a young child who is neither bright nor slow, as is stated in Yoreh Deah 274[:4] with regard to a Torah scroll.

Similarly, the initial and preferred option is to be careful not to leave more than a hairsbreadth between letters. After the fact, however, [the passage] should not be disqualified unless the word appears to be divided in two to a young child who is neither bright nor slow, as is stated (ibid.) with regard to a Torah scroll.

Some authorities maintain that one must leave an empty space the [height] of a line [of script] between the lines [of a passage], as [is necessary] with regard to a Torah scroll. Others maintain that one must make a point of doing this only [when writing] a Torah scroll. In fact, the common practice of scribes is [to follow the latter view] and not make a point of doing this [when writing] tefillin and mezuzos.

47 Whether [one is writing] a Torah scroll, tefillin, or mezuzos, he should not leave any more space between one verse and another than [is left] between the words within each verse. [The rationale is] that we are not knowledgeable with regard to the division of the verses that was transmitted to Moshe at Sinai.159

Nevertheless, if one did leave [space between verses, the passage] is not disqualified unless one left the amount of space that is required to be left between one passage and another. The initial and preferred option for the latter space is the width of the three words אשר אשר אשר, measured according to the style of script used within these passages. After the fact, the minimal acceptable space is the width of nine small letters, i.e., [nine times the width of] the letter yud, measured according to the style of script used within these passages, as stated in Yoreh Deah 275[:2].160

48 The initial and preferred option is to be careful to write all the lines of equal length, and not have one line indented within the margin or protruding beyond it by even a single letter. If, however, [a scribe] happens to reach a five-letter word at the end of a line, and he cannot write it within the line using his present style of writing, he may write three letters of it within the margin and two letters outside the margin.161 He may not, however, write two letters within the margin and three letters outside it.

Even if [the last] word [on a line] has eight letters, one may not write five within the margin and three protruding, even though the greater part of the word is within the margin. Moreover, two letters which form a word may not be written outside the margin.162

All the above applies as the initial and preferred option. After the fact, even if [the scribe] wrote a word comprising many letters beyond the margin, this does not disqualify [the passage].

49 Initially, the letters of G‑d’s name — whether the unique name,163 or any of the other Divine names which may not be erased, or even the letters which are suffixed to G‑d’s name164 — must all be within the margin, and should not protrude at all.

After the fact, [such passages] are acceptable for tefillin (since they cannot be corrected, lest the letters be written out of sequence). With regard to a Torah scroll, by contrast, [one should not rely on this leniency, for] one may scrape away all the preceding lines [of that column] if they do not include any of the names of G‑d that cannot be erased, and then extend them so that they will be even with the name of G‑d that protrudes.

50 Since the four passages of the arm-tefillin are written on one parchment scroll,165 they are bound by the laws pertaining to Scriptural passages which are pesuchos (“open”) and setumos166 (“closed”). For when a space [at least] nine letters wide is left at the end of the last line of the first passage, and the next passage starts at the beginning of the [first] line in the second column, that next passage is called pesuchah.167 This is the same as with a Torah scroll: A passage which starts at the beginning of a column is called pesuchah when it is preceded by a space [at least] nine letters wide at the end of the last line of the previous column.

In the arm-tefillin, therefore, every passage must be written as it appears — whether pesuchah or setumah — in a Torah scroll. This means that a space [at least] nine letters wide must be left at the end of the last line of the passage Kadesh [which is the first passage of the tefillin], so that the passage Vehayah ki yeviacha which starts at the beginning of the first line of the second column will be pesuchah, as it appears in the Torah. If it is written as setumah, it is invalid.

The concepts of pesuchah and setumah do not apply to [the beginning of] the passage Kadesh, since nothing is written before it in the tefillin. It therefore starts at the [very] beginning of the first line [of the first column]. Although it is written as pesuchah in the Torah, no space is left before it [in the tefillin] because nothing else is written before it — and no passage could have a more “open” beginning than one that has no other script to “close” [the passage that precedes] it.

At the end of the last line of the passage Vehayah ki yeviacha, one should leave a space [at least] nine letters wide so that the passage Shema, which starts at the beginning of the first line of the third column, will be pesuchah, as it appears in the Torah. If it is written as setumah, it is invalid.

The passage Vehayah im shamoa, by contrast, is setumah in the Torah, and thus should also be setumah in tefillin. If it was written as pesuchah, some authorities deem it invalid, though after the fact others deem it valid [for the following reason].

Although in the Torah it is setumah, nevertheless, in the Torah does not follow directly after the passage Shema at all.168 Hence the fact that [with regard to tefillin] the space between it and the [preceding] passage Shema is pesuchah, and not setumah like the “closed” space before it in the Torah, is not considered a deviation — because in the Torah, too, the [beginning of the] space before it is not “closed” by the preceding passage Shema. [On the contrary,] in the Torah there is considerable space between the passage Shema and [the later passage Vehayah im shamoa].169

With regard to the passage Shema, by contrast, the fact that it does not [follow directly after] the passage Vehayah ki yeviacha in the Torah does not detract from the necessity to write it as pesuchah. The halachah rules according to this view.

If, however, no space at all was left between the passage [Vehayah im shamoa] and the [preceding] passage Shema on any line in their columns, neither in the column of [Vehayah im shamoa] nor in the column of Shema — i.e., [the passage Vehayah im shamoa] started at the beginning of the first line of [its] column, and no preceding space whatsoever was left on that line nor at the end of the [last] line of the passage Shema, or even if a space was left but it was not of the nine-letter width that separates passages in the Torah — this is considered a (total deviation) from the manner in which it is written in the Torah, and it is invalid even after the fact. Rather, there should be a space which is setumah left before it, as is left before all passages which are setumos in the Torah.

At the conclusion [of the passage Vehayah im shamoa], however, it is not necessary to leave any blank space. This in fact is the custom — to conclude the words al haaretz at the end of the last line [of the passage Vehayah im shamoa].

51 A passage that is setumah in the Torah can appear in any of four possible forms. One of them is valid according to all opinions, and concerning three, there is a difference of opinion between Rambam and Rabbeinu Asher, each of them disqualifying the forms that the other approves.

If one completed a passage at the beginning of one line, writing at least one word of that passage at the beginning of that line, left a space [at least] nine letters wide on that line and then began the following passage on the same line, writing at least one word of the latter passage on this line, the passage which follows is considered as setumah according to all opinions.

If, however, one concluded a passage at the end of a line or within nine letters of the end of the line, and began the next passage on the following line after leaving a space of nine letters, the passage which follows is considered pesuchah by Rabbeinu Asher. Rambam, by contrast, will deem such a passage too as setumah, whether it [began] on the next line of that column or at the beginning of the next column — unless they were separated by an entire blank line, either at the foot of that column or at the beginning of the next column, as will be explained.

Similar concepts apply if one completed a passage and at its conclusion left a space of nine letters’ width at the end of that line, and did not start the next passage at the very beginning of the next line, but rather left a little space — even less than nine letters — and then started the next passage. Rambam would deem the [latter] passage as setumah. For he considers as setumah any passage that does not start at the beginning of its line when there is a space before it which is nine letters wide, whether on that line or on the previous line. [This applies] even if it starts at the head of the following column. According to Rabbeinu Asher, by contrast, as long as there is a nine-letter space, whether at the beginning of the [new] line or at the end of the [previous] line, the [second] passage is considered as pesuchah.

If one concluded a passage at the end of a line or less than nine letters’ width from the end of the line, then left one complete line blank, and started the next passage at the beginning — or within nine letters of the beginning — of the third line, the passage to follow would be considered by Rabbeinu Asher as setumah. For he considers the space of the blank line between the two passages as if it were in the middle of the [same] line [as] the two passages, thus making the latter passage setumah. According to Rambam, by contrast, [such a passage] would be considered as pesuchah, regardless of whether the end of the first passage and the beginning of the second passage are in the same column or in two separate columns and [regardless of whether] the empty line is at the end of the first column or at the beginning of the second column.

52 Now, when [writing] tefillin, it is impossible to write the passage Vehayah im shamoa in a form that all authorities will consider as setumah, because nothing else is written before it on the line on which it begins. For the passage Shema ends in its own column, and [Vehayah im shamoa begins] a new column. If one were to write it at the beginning of the line at the head of the column and leave a space of nine letters at the end of the passage Shema, all authorities will consider it as pesuchah, whereas it is setumah in the Torah.

If [on the other hand] one does not leave a space of nine letters [before it], it is certainly not [written] as it is [written] in the Torah, for if it is not preceded by a space of nine letters it is not considered a distinct passage at all.

Thus the only alternative is to write [the passage] as setumah in a manner which [conforms to the view of either Rabbeinu Asher or Rambam;170 for example,] to follow Rabbeinu Asher’s view in contrast to that of Rambam, by leaving one line blank and beginning the passage on the second line, or to write it in a manner that Rambam — but not Rabbeinu Asher — would consider setumah, i.e., not to start it at the beginning of the line, but first to leave a space of nine letters at the beginning of the line. This would render the passage as setumah according to Rambam.

Nevertheless, [when doing so,] one should be careful not to leave a space of nine letters at the end of the passage Shema.171 [When writing] a Torah scroll, if it happens that [a space] of nine letters was left at the end of a line and there is not enough additional space to write another word from the next passage on this line, it is still considered as a passage which is setumah if one leaves a small space at the beginning of the following line. For a passage which is pesuchah always starts at the beginning of a line and, according to Rambam, a passage which does not start at the beginning of the line is considered as setumah. Nevertheless, [when writing tefillin, one should not rely on this leniency].

[The rationale is that] the space at the end of the line indicates a passage which is pesuchah, since when there is no other space on the next line, the passage which follows it will be pesuchah. And if there is no space of this measure at the end of the [preceding] line, this indicates that the following passage will be setumah.

Thus with regard to tefillin, where the scribe has the option of writing the passage as he desires, he should conclude the passage Shema at the end of the line or within less than nine letters of the end of the line, to indicate that the passage which follows it is setumah. It is only with regard to a Torah scroll that we apply [the leniency of] considering a passage as setumah when [the scribe] leaves a space of nine letters at the end of the line in the preceding passage because the following passage does not start at the beginning of a line. For it is forbidden to extend or shorten the letters to [adjust] the space between passages, as stated in Yoreh Deah 273[:2].172 But if it happens that [a scribe] writing tefillin leaves a space of nine letters at the conclusion of the passage Shema, he does not have to leave a space of nine letters at the beginning of the following passage. All that is necessary is to leave a small space.

If, however, [a scribe] did not leave a space of nine letters, neither at the conclusion of the passage Shema nor at the beginning of the following passage, [the passage is invalid. This applies even when,] should one combine [the space at the end of the passage Shema and the space at the beginning of the passage Vehayah im shamoa], there would be a space totalling nine letters or more. The two cannot be combined. Thus the passage is not at all considered [even] as pesuchah,173 because there is no space of nine letters.

There is, however, a view that [the two spaces] can be combined. Thus the passage which follows would be considered as setumah according to Rambam, and perhaps also according to Rabbeinu Asher. Hence, according to this opinion, it would be initially preferable to write a passage in this manner, so as to satisfy both approaches.174

Nevertheless, one should be mindful of the first opinion and not place himself in a situation in which it is possible that one’s tefillin are invalid. For according to the first view, which holds that in this instance there is no [acceptable] passage whatever, [the tefillin] are invalid according to all opinions, even after the fact. If, by contrast, [the passage Vehayah im shamoa] is written as pesuchah, the halachah follows the view that after the fact this is acceptable.

53 Since it is impossible to write [the passage Vehayah im shamoa] as setumah [in a form] that all authorities accept, it is customary in many countries to write it [in a form that would be considered] as setumah according to Rambam, for the halachah follows his approach. Even with regard to a Torah scroll, when it is impossible to write [a passage in a manner which satisfies all opinions, the approach of Rambam is followed], as stated [by Rama] in Yoreh Deah 275[:2].

Nevertheless, in these countries it is customary to write [this passage] in a form that is pesuchah like the passages which precede it, for there is a view that permits one even initially to write [this passage] as pesuchah, since in the Torah it is not located next to the passage Shema. Even though the halachah does not follow this view, since after the fact [a passage written in this manner] is valid according to all views, and it is impossible to write it as setumah in a manner that will be accepted by all authorities, the later authorities rely on this opinion so as not to negate the existing custom175 and thereby cast aspersions on [the tefillin of] past generations.

54 All of the above applies [only] to arm-tefillin. With regard to the head-tefillin, by contrast, since its passages are written on four separate scrolls, there is no need to be concerned whether they are written as pesuchah or setumah. The custom is, however, to write [them with these laws in mind].

55 Our Sages received the tradition176 that the word totafos, by which the Torah names the head-tefillin, refers to four compartments of leather for the four passages, each in a separate compartment.

The [four] passages of the arm-tefillin, [by contrast, should be placed in] a single leather compartment. [This is implied by the verse,177] “And it shall be as a sign for you [on your arm].” [It is as if G‑d said:] “I prescribed one sign, not more than one.”

56 The leather for the compartments,178 both for the arm-tefillin and for the head-tefillin, must be from a kosher species of domesticated animal, beast, or fowl. Even one which died without ritual slaughter or which was treifah [is acceptable], provided [the species] is “permissible [as food] in your mouth.”179

The compartments may be made of parchment, for this, too, is called leather. One authority, however, holds that the compartments should not be made of parchment unless it was first saturated in a solution made of tree bark (which is called lye), and which is used to process leather. Initially, it is advisable to respect this opinion.

[The compartments] may also be made of the hide of an animal fetus. This, too, is considered as leather in this context, although with regard to the laws of ritual impurity it is considered “meat.”180 Similarly, the [tefillin] straps may be made of parchment or the hide of an animal fetus.

According to certain authorities it is preferable to use the hide of an animal fetus, for it was never involved in a sin since it had never mated. There is also an authority who states that after the hide of an animal fetus, [the order of preference is as follows]: the skin of a fowl, the skin of a wild beast, the skin of an animal which was slaughtered in a kosher manner, and the skin of an animal that died without ritual slaughter or which was treifah.

57 The leather used for the compartments must be processed with the intent that they be used for the mitzvah,181 as is required for the parchment used for the passages.182 Some authorities, however, maintain that the two are not alike, for the holiness [of the tefillin abides] mainly in the passages, whereas the compartments are their accessories. [Hence, it is argued,] they need not be processed for the sake of the mitzvah.

The [latter] opinion may be relied upon when there is no other alternative, so that one’s observance of the mitzvah of tefillin not be suspended. [Indeed,] one may recite a blessing over such tefillin until one finds others.183

58 The four compartments of the head-tefillin must be made of a single piece of leather, as implied by the verse which refers to them as “a reminder between your eyes.” [It is as if G‑d said,] “I prescribed one reminder, not two or three.”184

Pieces of leather which are sewn together are considered as a single piece of leather. (The same applies if they are attached with glue.)185 Nevertheless, it is proper and preferable to make them of actually one piece of leather;186 i.e., a piece of soft calf-hide is fitted into the hollows of a mold and left to dry. Thus, when the mold is removed, the compartments will have been formed as a natural consequence.

59 It is a halachah transmitted to Moshe at Sinai that [the cubes housing the passages of] both the arm-tefillin and the head-tefillin should be exactly square;187 i.e., their length should equal their width, so that their diagonal will be approximately two-fifths longer than their length.

As will be explained,188 the titura [which is the projecting base surrounding each of these cubes] must also be exactly square. The stitches [on all four sides] must also describe an exact square189 whose length equals its width. Moreover, the cube itself must be square; i.e., its length and width must be the same (throughout its height). Its height, however, may exceed its length and width.190

If [the tefillin] were initially square and then the measure of the square was marred in one of the above places,191 they must be corrected. If they are not corrected, they are invalid, for [the source for this requirement] is a halachah transmitted to Moshe at Sinai.192

With regard to the head-tefillin, all four compartments must together — not separately — constitute a square.193 Initially, it is preferable that all four compartments should be made the same size; one should not be wider than the other.

60 It is a mitzvah to make the leather of the compartments black,194 as the straps are. Though the halachah transmitted to Moshe at Sinai requires that the straps [must be black],195 whereas the compartments are not [included in this requirement], the mitzvah of tefillin requires that everything be black.

61 The grooves between the compartments must each reach as far [down] as the stitches.196 [Nevertheless, the tefillin] are valid even if the grooves do not reach there, provided they are discernible above, so that it will be obvious to all that there are four [separate] compartments.197

When does the above statement [concerning validity] apply? When the internal partitions — which separate one compartment from the other and which extend downward for the full [height] of each parchment scroll — are made of the same piece of leather as the outer walls [of the tefillin]. However, if [the compartments] were not fashioned in this way, or they were initially, but the leather [partition] between passages was later torn, the [tefillin] are invalid, as will be explained in sec. 33[:1].

62 There is no specified measure for the height of the tefillin; they may be made as large or as small as one desires. Sages and heads of Rabbinic courts customarily make their head-tefillin more than three fingerbreadths high,198 for they are worn visibly, and it is praiseworthy that they be visible, as was explained in sec. 27[:23].

[Lesser] sages and students, and [in particular,] a student in the presence of his mentor, for whom it would not be respectful to uncover his tefillin in his [mentor’s] presence, wear small tefillin, a fingerbreadth-and-a-half high, under their hats. This expression of respect indicates that they do not equate themselves to their mentors in [their] presence. [To wear tefillin as large as their teachers’] would appear as brazenness, for tefillin are an expression of glory and greatness, as indicated by the verse,199 “And all the nations of the world will see that the name of G‑d is proclaimed upon you and they will fear you” — on which our Sages comment,200 “This alludes to the head-tefillin.”

63 Similarly, there is no specified measure for the [combined] width and length of the [four] compartments [of the head-tefillin]. There is, however, a required measure for the length and width of the base: by initial preference it should be larger than a fingerbreadth at the very least.

[This is derived as follows:] Our Sages taught that201 “there is place on the head to put on two tefillin.” If it were acceptable [to use tefillin which were] a fingerbreadth [wide], there would be space [not only for two tefillin, but] for three.202

The Geonim prescribe that the initial and preferred option is that the compartments of the head-tefillin [total] two fingerbreadths by two fingerbreadths, though if smaller, they are not disqualified. The initial preference for the cube of the arm-tefillin is that it be a fingerbreadth by a fingerbreadth.203 The term “fingerbreadth,” unless otherwise qualified, always refers to [the width of] the thumb. One should be mindful of this [preference], though this is not the [popular] custom.

64 It is a halachah transmitted to Moshe at Sinai that a protruding shin be formed from the creases in the leather of the head-tefillin on both the right and left [outer] sides [of the compartments].204

The Geonim received the tradition that the shin to the right of the person wearing [the tefillin]205 should have three heads, and the shin to his left should have four heads206 — though reversing this does not disqualify [the tefillin].

The yud-shaped heads of each of these letters shin must extend downward, reaching the base of the shin, and the point at the base of the shin should preferably reach [down] as far as [the level of] the stitches [at the base of the tefillin]. Each shin should not, however, be extended downward excessively; rather, the point at its base should preferably be visible above the [level of the] stitches.

65 Some authorities hold that the protruding shin must result [directly] from the creases of the leather.207 [These authorities argue that] if the letter [protruding from the leather] resulted [not only from an internal convex mold but in addition] from [the reciprocal pressure of] a [concave] mold [outside the compartment], the obligation would not be fulfilled, because this process resembles chak tochos.208 It is true that the shin-shaped protrusions on the [internal] mold press the leather outward and engrave their form upon it, and thus it is the [protruding] body of the letter that is being engraved, and not [the depressed intervening spaces of] its interior. Nevertheless, the leather is soft and is also pressed around the shin-shaped protrusion on the [internal] mold by the [external] mold in which the concave shin is engraved. [The notched surface of the external mold] thus stamps the depressed spaces of the shin into the leather. Hence, [it is argued,] this [bi-directional process] constitutes an instance of chak tochos.209

Other authorities, however, maintain that the factor of chak tochos need not be considered with regard to this shin. [They argue that] chak tochos is a disqualifying factor only with regard to [writing] the passages of tefillin and mezuzos, and similarly with regard to a Torah scroll and a bill of divorce, for [in these cases] the Torah explicitly specifies “writing” — and this excludes engraving.210 By contrast, the halachah transmitted to Moshe at Sinai concerning this shin does not specify “writing.”

And indeed, the common practice211 is to imprint this shin using [two] mold[s]. Nevertheless, it is advisable to respect the first view.

66 It is a halachah transmitted to Moshe at Sinai that the arm-tefillin and the head-tefillin each have a titura [or “base”].212 Titura is Aramaic for “bridge,” implying that there be a leather pad at the base [of the tefillin] to cover the opening of the compartments. It thus appears like the plank of a bridge.

Similarly, it is a halachah transmitted to Moshe at Sinai that the arm-tefillin and the head-tefillin each have a maabarta [or “passageway”];213 i.e., the leather of the titura should extend and project at one side. This projection should be of two layers, allowing the strap to pass through it, hence the name “passageway.”214

Since it is a halachah transmitted to Moshe at Sinai that the titura at the base of the tefillin should be exactly square,215 the maabarta must be cut [as a marginal indent] at both sides (at the point at which the square of the titura ends) so that the maabarta will not be as wide (at that place) as the titura. In this way, the square shape of the titura will be distinct.

67 Before each parchment scroll is inserted in its compartment it must be rolled up, starting from the end, just as a mezuzah is rolled from the word echad towards the word Shema,216 and not in the opposite direction, for the reason explained in Yoreh Deah 288.217

68 Some authorities hold that it is a halachah transmitted to Moshe at Sinai that [a strip of] cloth or parchment be bound around [each of] the passages218 [before it is placed in its compartment],219 and that otherwise they are invalid. Other authorities differ.

As an initial preference, it is customary to bind [a strip of] parchment around [each scroll]. If this was not done, the [tefillin] are nevertheless valid after the fact if no others are available.

Some people make a point of binding [the scrolls] only with parchment [of a kind] that would be valid for making the compartments or the straps. [The rationale is that] since binding this [strip of parchment] is a halachah transmitted to Moshe at Sinai, the laws governing it are identical with those governing the compartments and the straps. Hence, it must derive from a species [of animal] which is permissible as food,220 and not from a forbidden species. If a cloth [is used], however, [this] restriction does not apply and is thus not an issue for concern.221

69 It is a halachah transmitted to Moshe at Sinai that strands of hair from a kosher animal or beast be wound around each [scroll; this is done before it is inserted in its compartment].222 Some authorities hold that the hair should be wound around the [scroll] itself and the [strip of] parchment should then be bound around it. Others hold that first the [strip of] parchment should be bound around the [scroll] and then the hair should be wound around the parchment.

The prevailing custom is to satisfy both opinions. First the hair is wound around the [scroll], then [a strip of] kosher parchment is bound around it, and then the hair is wound around [the parchment].223

The hair of a calf is customarily used,224 so that one recall the sin of the [Golden] Calf and not transgress. If a calf’s hair is not available, the hair of a cow or an ox should be used. The hair should first be washed thoroughly clean.

The Sages of the Zohar225 caution that these hairs should protrude from the compartments of the head-tefillin, though [no more] than the length of a grain of barley. Some sources state that they should protrude from [the compartment containing] the passage Kadesh, while others state that they should protrude from [the compartment containing] the passage Vehayah im shamoa.226 It is preferable that they protrude from the side [of the compartment containing] the passage Vehayah im shamoa that is [closer] to [the compartment containing] the passage Kadesh.

70 The initial and preferred option is that each of the inscribed scrolls be inserted vertically in its compartment, as a Torah scroll stands in the ark. If the scrolls were laid in the compartments horizontally, the [tefillin] are nevertheless valid.

Care should be taken not to place the passages in the compartments upside down.227 Instead, the first line of the passage should be at the top of the compartment and the bottom line should be near the lower, open end of the compartment.

Initially, one should ensure that the beginning of each passage is oriented towards the side [of the compartment] to the right of a [hypothetical] reader228 who is facing the person wearing [the tefillin]. If [such a person] would open them and read the passages, they would be set out before him in their required order. [This is significant,] for we find that the order in which [the passages] are placed in their compartments is determined by their respective positions to the right of the reader, so that he can [hypothetically] read them in order: first the passage Kadesh to his right, then the passage Vehayah [ki yeviacha], and the other passages to his left, as will be stated in sec. 34[:1].

71 If all four passages [for the head-tefillin] were written on a single piece of parchment, they are valid, even if no space was left between them — provided they are placed in four [separate] compartments, each one lying [horizontally] in its compartment. [However,] a thread or cord must be placed between the adjoining compartments, to separate them from each other.

Some authorities maintain that [the Sages] require this distinction only if all four passages are written on a single piece of parchment, whereas if they are written on four separate pieces of parchment as dictated by law, there is no need whatsoever for a thread or cord. Others, however, maintain that even if [the passages] are written on four separate pieces of parchment, a thread or a cord must [separate] one compartment from the other.229 This indeed is the common practice: by initial preference, the thread used to stitch [and close the head-tefillin] is [first] passed between the adjoining compartments. If, however, this is not done, [the tefillin] are nevertheless valid, for the halachah rules according to the former view.

72 The laws which apply to [the passages of] the head-tefillin — that each of the scrolls should be bound around with hair and [a strip of] parchment, that they should each be rolled from end to beginning, and that they should be placed vertically in their compartments — also apply to the passages of the arm-tefillin, except that by initial preference [all four passages of the arm-tefillin] should be written on a single piece of parchment, for the reason stated above.230

If a [scribe] wrote [the passages for the arm-tefillin] on four separate pieces of parchment and placed them in four separate compartments, like the head-tefillin, he has fulfilled his obligation,231 provided he drapes a piece of leather over them so that they appear as one compartment. For [the arm-tefillin] must appear as one “sign” from the outside, as stated above.232

Needless to say, if a [scribe] wrote [the passages for the arm-tefillin] on four separate pieces of parchment and placed them in one compartment, he has fulfilled his obligation. There is no need to glue the pieces of parchment together. Nevertheless, the custom is to glue them together, care being taken to use kosher glue, i.e., glue made from a kosher animal, [from a species whose flesh] is permissible as food.233

73 If the compartments were coated with gold, or with leather that may not be used for making compartments, e.g., leather from a non-kosher species [of animal], [the tefillin] are invalid.234

74 It is a halachah transmitted to Moshe at Sinai that the tefillin be [closed by being] stitched with sinews from a kosher species of animal or beast.235

([It is customary] to take white sinews from the heel236 of an animal or beast. If they are stiff, they should be softened with stones until they resemble [threads of] flax, and then they are twined and spun.)

A certain scholar cites a tradition he received from the Geonim that, as an initial preference, it is a mitzvah to stitch [tefillin] with the sinews of an ox. It is advisable to defer to this view.

75 One should not go to non-Jew premises to purchase sinews if it is known that some [gentiles] take them from non-kosher species [of animal], even though most take them from kosher species.

[The rationale is that] since the [purchaser] is going to the non-Jew while the latter is in residence at home, [the non-Jew is considered kavua — “in a fixed position.” [In such an instance] we do not follow the majority, because of the principle that “whenever something is kavua, its status is considered as half and half.”237

If, however, a non-Jew brings [sinews] to the market [to sell], it is permitted to purchase [them] from him. [The rationale is that] since he has left the place where he is “in a fixed position,” we follow the principle that “whenever an item has been separated [from its accustomed place], we assume that it is one of the majority that has left its place.”238

If it is not known that some gentiles take sinews from a non-kosher species [of animal], it is permitted to purchase them in all instances. The reason: Most of the animals whose meat is eaten239 are of a kosher species, and sinews are not taken from a species which is not eaten, for a non-Jew will certainly not kill his animal for the sake of its sinews.

The above applies to sinews which were not spun. If the sinews have been spun, it is forbidden to purchase them from [gentiles] in all instances, for since [the sinews] are required by a halachah transmitted to Moshe at Sinai, they must be processed [lit., “made”] for the sake of the mitzvah. To cite a parallel: The blackening of [tefillin] straps must be done for the sake of the mitzvah, because it is a halachah transmitted to Moshe at Sinai that [the straps be] black, as stated in sec. 33[:4-5]. The sinews are “made” by spinning, which must therefore be done for the sake of the mitzvah. (To cite [another] parallel: The threads for tzitzis are “made” by spinning, which must therefore be done for the sake of the mitzvah.)240 [Hence the sinews may not be spun by] gentiles, for they will not perform an activity for the sake of a mitzvah.

76 There is an authority who maintains that strings made of parchment are considered equivalent to sinews and are valid for use in stitching Torah scrolls and tefillin. In places where there are no sinews at hand one may rely on this view and [close tefillin] by stitching with these strings until sinews become available, so that one’s observance of the mitzvah of tefillin not be suspended.

77 There is a tradition received by the Geonim that one should sew [and close the tefillin with] twelve stitches as an allusion to the Twelve Tribes, three stitches on each side [of the compartment]. With every single stitch, the thread must pass over both [the top and the bottom of the base].241 If one made fewer than twelve stitches or more than twelve, [the tefillin] are not disqualified.

The stitching need not be within the actual compartment[s]; [it is sufficient that it be] close to the compartment[s]. Nevertheless, the leather [extending beyond the lower edge] of the compartment must be spread outward and be sandwiched between both layers of the base, [at least] as far outward as the stitching, so that the stitches also penetrate the leather extending from the compartment (which is considered part of the actual compartment).242 Thus the body of the actual compartment is stitched into the base.

If this is not done, [the tefillin] are invalid, for it is a halachah transmitted to Moshe at Sinai that tefillin be [closed by being] stitched,243 and it is the compartments that are considered the tefillin [proper], not the base. This is also why the actual compartments must be made square, not only the base, for the Sages taught: It is a halachah transmitted to Moshe at Sinai that tefillin must be square,244 and it is the compartments that are referred to as the tefillin [proper], whereas the base is called the base of the tefillin.

78 There are those who are stringent and remove the thread used [to close the tefillin by] stitching if it snaps in the middle. They remove it entirely and stitch [the tefillin] anew with another thread.245 [The rationale is that] since it snapped in the midst of the sewing, it is surely not strong and will ultimately snap — and [such a thread] is considered as if it has snapped already, as if it does not exist.

If, however, the thread was short at the outset and did not snap, the sewing may be resumed with another thread. It is not necessary that a single thread be used for all the stitches.

79 The knot with which tefillin are tied is a halachah transmitted to Moshe at Sinai.246 After inserting the straps in the passageway, one should tie a knot which resembles a dalet for the head-tefillin and one which resembles a yud for the arm-tefillin. This — together with the shin [on the compartment] of the head-tefillin — completes the letters shin-dalet-yud, [spelling out] the name of G‑d.247

It is advisable to make a point of completing the letters of the Divine name in the correct order, i.e., that the shin should be made first, then the knot of the head-tefillin should be made, and then the knot of the arm-tefillin in the order of the letters of [this Divine] name.

The knot of the head-tefillin need appear as a dalet only from the outside; it does not also need to appear thus on the underside that faces the head. There are those, however, who follow the custom of making it appear as a dalet from both sides.248

80 It is customary to cover the compartment of the arm-tefillin with a piece of leather over the width of the arm, because the arm-tefillin is tall and narrow, and [can be] damaged by the clothes under which it is placed. It is therefore customary to cover it with this strip [of leather] in order to strengthen it. In some communities [this] is not the custom.249